Dementia, the loss of cognitive and behavioral abilities, is one of the major causes of disability and dependency among older people worldwide. Neurodegenerative diseases, such as Alzheimer?s disease and Lewy body dementia (including dementia with Lewy bodies and Parkinson?s disease dementia), are the most common causes of dementia. Currently, there are several diagnostic and therapeutic challenges for these diseases. Validated biomarkers of neurodegeneration, especially at premanifest stages, do not exist. Further, there is no available treatment to slow down or arrest the progression of disease. Thus, there is a huge need of biomarkers able to detect neurodegeneration early in the cascade. Brainstem imaging holds great promise in assessing Lewy body dementia in elderly humans, when Lewy bodies (i.e. neuro-degeneration due to the accumulation of alpha-synuclein protein in the brain) are expected to affect the brainstem and the olfactory bulb, before spreading to other brain areas during symptomatic stages. Brainstem-based biomarkers can fill the currently existing holes in diagnosing the early synucleinopathy stages when treatment can be most effective in delaying the development of full- blown neurodegeneration. However, existing imaging methods are incapable of resolving details of tiny deep brainstem nuclei in living elderly humans, limiting the development of brainstem-based biomarkers of premanifest Lewy body dementia. To fill this gap, the central aim of the proposed research is to create a probabilistic atlas of twenty arousal and motor brainstem nuclei in healthy elderly subjects by ultra-high field (7 T) MRI. The atlas will be used to evaluate brainstem-based biomarkers of premanifest Lewy body dementia (LBD) assessing the integrity of brainstem nuclei microstructure and connectivity pathways. To do so, we will map the brainstem nuclei atlas to both advanced (e.g. 7 T) and clinical (e.g. 3 T) neuroimages of controls, premanifest LBD, as well as de novo manifest LBD; crucially, we will test the hypothesis that brainstem microstructure and/ or connectivity pathway are altered in premanifest LBD compared to controls, and that these changes become stronger and affect more brain regions in de novo manifest LBD. Thus, our project will provide two important new tools, a structural atlas of brainstem nuclei and brainstem-based biomarkers of premanifest LBD. The generated brainstem nuclei atlas will be a useful tool in research and clinical studies of elderly populations able to automatically identify the location of brainstem nuclei in both advanced and clinical MRI of elderly subjects. The developed brainstem- based biomarkers of premanifest LBD will improve our understanding and the diagnosis of early stage Lewy body dementia and the development of better treatment for early stage patients before they manifest cognitive and behavioral symptoms.